汽车工程 ›› 2021, Vol. 43 ›› Issue (9): 1328-1335.doi: 10.19562/j.chinasae.qcgc.2021.09.009

• • 上一篇    下一篇

基于车辆模型紧耦合的封闭园区车辆定位方法

秦晓辉,王哲文,庞涛,史维清,孙宁,胡满江   

  1. 汽车车身先进设计制造国家重点实验室,湖南大学机械与运载工程学院,长沙 410082
  • 收稿日期:2021-02-25 修回日期:2021-05-23 出版日期:2021-09-25 发布日期:2021-09-26
  • 通讯作者: 孙宁
  • 基金资助:
    湖南省重点领域研发计划(2019GK2151);汽车车身先进设计制造国家重点实验室(61775006);中央高校基本科研业务费资助

Vehicle Positioning Method Based on Tight Coupling of Vehicle Model in Enclosed Environments

Xiaohui Qin,Zhewen Wang,Tao Pang,Weiqing Shi,Ning Sun,Manjiang Hu   

  1. State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body,HNU College of Mechanical and Vehicle Engineering,Changsha 410082
  • Received:2021-02-25 Revised:2021-05-23 Online:2021-09-25 Published:2021-09-26
  • Contact: Ning Sun

摘要:

自主定位是自动驾驶车辆的一项基本能力,全球导航卫星系统(GNSS)可在开阔环境提供定位解决方案,然而在封闭园区环境如港口或工业园区等,高密度的植被和建筑等环境因素会导致GNSS信号不稳定,从而影响定位精度,对自动驾驶系统的安全造成严重威胁。为解决这一问题,本文中提出融合激光雷达(LiDAR)和惯性测量单元(IMU)的自动驾驶定位方法,通过引入车辆运动学模型以约束车辆位姿优化方向,同时采用模块化设计思路构建系统残差并基于紧耦合方法联合优化获得车辆准确位姿。试验结果表明,所提出的方法能够提高弱GNSS信号环境中自动驾驶车辆的定位精度和鲁棒性。

关键词: 自动驾驶, 自主定位, 紧耦合, 车辆运动学模型

Abstract:

Autonomous positioning is a fundamental capability of autonomous vehicles. Global navigation satellite system (GNSS) can provide positioning solutions in open environments, however, in enclosed environments, such as ports or industrial parks, environmental factors such as high?density vegetation and buildings will lead to instability of GNSS signals, thus affecting positioning accuracy and posing a serious threat to the safety of the automatic driving system. In order to solve this problem, this paper presents an automatic driving positioning method that integrates light detection and ranging (LiDAR) and inertial measurement unit (IMU). A vehicle kinematics model is introduced to constrain the orientation of vehicle poses optimization, at the same time, the system residuals are built by using the modular design idea, which are optimized jointly based on tight coupling method to obtain the accurate pose of the vehicle. The experimental results demonstrate that the proposed method can improve the positioning accuracy and robustness of autonomous vehicles in weak GNSS signal environments.

Key words: automatic driving, autonomous positioning, tight coupling, vehicle kinematics model